If you think Snowden is intentionally cooperating with those governments, you have the burden of explaining why he went public, instead of just moving into his dacha and enjoying his ill-gotten gains.

If you think those governments have tried to get the data without his knowledge, bear in mind that he's technically sophisticated and it's inconceivable that he didn't encrypt his drive.

If you think the Russians are employing rubber hose cryptanalysis, bear in mind that he is still in contact with several western journalists, and it's very likely that they have prearranged codes to let him communicate such a message.

True. This is actually one of the more legitimate things the feds are doing. It has a clear connection to anti-terrorism efforts, and I really don't think people who are using banks to transfer money have much expectation of privacy, as they do when talking on the phone, emailing, etc.

Consider: in the US, you generally get only $20 bills from the ATM. You go and spend these with retailers who usually get nothing larger than $20 bills. The $20 bill you spent thus is not recycled as change, and it gets deposited at the end of the day. That makes a nice short loop that's easy to analyze.

So if you want to buy something and not make it too easy to track the bills, use denominations under $20.

Now, as for things like making a drug buy, where you really wouldn't want the gov't to track it, that's probably less of a problem because those people tend not to use banks so much...

Good point. I see two reasons for "the authorities" to be more worried about 3D guns. First is (I would guess) CNC is fairly mature and stable, while 3D printing is undergoing rapid development, becoming cheaper, more versatile, more accessible. Second is that a CNC machine is something of a niche application, not interesting to many people, whereas 3D printing can make a much wider variety of things and is therefore potentially of interest to more people. Thus, 3D printed firearms are potentially accessible to many more people, and, with future technology, easier to produce.

Yes, I just realized this. A properly written OS can periodically test the hardware RNG for reduced entropy. Let us suppose we can detect if the entropy has fallen below 32 bits. Then, whenever we are using the hardware RNG, we pessimistically assume that there are only 16 bits of entropy available per sample. Grab a bunch, run it through a good hash function, repeat, concatenate. You end up with as many bits of good random data as you need, and you XOR it with the random bits you got from other sources.

It seems that they did not control for exposure to fluoridated water. The article says "they had no data on the causes of missing teeth." It would be interesting to see if any clear results emerged from a study that did control for that.

However, given the level of entrenched interest in water fluoridation, I suspect it would be difficult to find funding for such a study, at least in the US.